On Sunday, 13. July 2003 04:05, Gregor Lawatscheck wrote: > At 19:17 12/07/2003, you wrote: > > > Then I played some recordings. Nothing happened. > > > >Second test completed successfully. My system was running more than 15 > >hours without sat signal. No problems. The frontend process consumed > >30-35% CPU power.
I just checked again with todays snapshot - the crash is still there... Having a revision 1.3 here, appearantly leaving the system alone keeps it healty, but if you have high sysload+recording+playback going on - that way the chips are stressed more i think (my card has an extra fan) - the crash happens soon. First one get's outcommand errors (resulting in long keay-reaction-times in playback mode) and error waiting command idle, later vdr-buffer usage goes to 100% and the ARM crashes. After watchdog reset the driver reloads but fails, because if some other program has still open /dev/video the necessary modules can not be realoaded and stay in kernel. PS: I verified - the one and only call to SetPIPs is with all zero's... maybe one should call that again for some reason? PPS: my kdvb-* threads are always low on CPU-usage... -- Info: To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as subject.
